[dpdk-dev] doc: update virtio doc
Checks
Commit Message
Add more explanations about vring size changes and virtio_header
size.
Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
---
doc/guides/nics/virtio.rst | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
Comments
> -----Original Message-----
> From: Yang, Zhiyong
> Sent: Wednesday, May 10, 2017 7:18 AM
> To: dev@dpdk.org; Mcnamara, John <john.mcnamara@intel.com>
> Cc: yuanhan.liu@linux.intel.com; maxime.coquelin@redhat.com; Yang, Zhiyong
> <zhiyong.yang@intel.com>
> Subject: [PATCH] doc: update virtio doc
>
> Add more explanations about vring size changes and virtio_header size.
>
> Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
> ---
> doc/guides/nics/virtio.rst | 13 +++++++++----
> 1 file changed, 9 insertions(+), 4 deletions(-)
>
> diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst index
> 91bedea..36ddd08 100644
> --- a/doc/guides/nics/virtio.rst
> +++ b/doc/guides/nics/virtio.rst
> @@ -41,8 +41,8 @@ The DPDK extends kni to support vhost raw socket
> interface, which enables vhost to directly read/ write packets from/to a
> physical port.
> With this enhancement, virtio could achieve quite promising performance.
>
> -In future release, we will also make enhancement to vhost backend, -
> releasing peak performance of virtio PMD driver.
> +In future release, we will also make enhancement to vhost backend.
> +Peak performance of virtio PMD driver will be released with every release
> version since DPDK 17.02 release.
Hi,
There have been great performance improvements in Virtio in the last few
releases. However, we can't really promise that is will improve in every
release. I'd suggest removing these two lines and the previous line
starting with "With this enhancement, virtio could ...".
>
> ...
>
> -* RTE_PKTMBUF_HEADROOM should be defined larger than sizeof(struct
> virtio_net_hdr), which is 10 bytes.
> +* RTE_PKTMBUF_HEADROOM should be defined
> + larger than sizeof(struct virtio_net_hdr), which is 10 bytes, when
> using non-mergeable.
> + larger than sizeof(struct virtio_net_hdr_mrg_rxbuf), which is 12
> bytes when mergeable or F_VERSION_1
> + is set.
>
All constants and struct names should be ```` quoted. Like this:
* ``RTE_PKTMBUF_HEADROOM`` should be defined larger than
``sizeof(struct virtio_net_hdr)``, which is 10 bytes, when using
non-mergeable or larger than ``sizeof(struct virtio_net_hdr_mrg_rxbuf)``,
which is 12 bytes when mergeable or ``F_VERSION_1`` is set.
John
Hi, John:
> -----Original Message-----
> From: Mcnamara, John
> Sent: Wednesday, May 10, 2017 5:58 PM
> To: Yang, Zhiyong <zhiyong.yang@intel.com>; dev@dpdk.org
> Cc: yuanhan.liu@linux.intel.com; maxime.coquelin@redhat.com
> Subject: RE: [PATCH] doc: update virtio doc
>
>
>
> > -----Original Message-----
> > From: Yang, Zhiyong
> > Sent: Wednesday, May 10, 2017 7:18 AM
> > To: dev@dpdk.org; Mcnamara, John <john.mcnamara@intel.com>
> > Cc: yuanhan.liu@linux.intel.com; maxime.coquelin@redhat.com; Yang,
> > Zhiyong <zhiyong.yang@intel.com>
> > Subject: [PATCH] doc: update virtio doc
> >
> > Add more explanations about vring size changes and virtio_header size.
> >
> > Signed-off-by: Zhiyong Yang <zhiyong.yang@intel.com>
> > ---
> > doc/guides/nics/virtio.rst | 13 +++++++++----
> > 1 file changed, 9 insertions(+), 4 deletions(-)
> >
> > diff --git a/doc/guides/nics/virtio.rst b/doc/guides/nics/virtio.rst
> > index
> > 91bedea..36ddd08 100644
> > --- a/doc/guides/nics/virtio.rst
> > +++ b/doc/guides/nics/virtio.rst
> > @@ -41,8 +41,8 @@ The DPDK extends kni to support vhost raw socket
> > interface, which enables vhost to directly read/ write packets
> > from/to a physical port.
> > With this enhancement, virtio could achieve quite promising performance.
> >
> > -In future release, we will also make enhancement to vhost backend, -
> > releasing peak performance of virtio PMD driver.
> > +In future release, we will also make enhancement to vhost backend.
> > +Peak performance of virtio PMD driver will be released with every
> > +release
> > version since DPDK 17.02 release.
>
> Hi,
>
> There have been great performance improvements in Virtio in the last few
> releases. However, we can't really promise that is will improve in every release.
> I'd suggest removing these two lines and the previous line starting with "With
> this enhancement, virtio could ...".
>
Ok.
>
> >
> > ...
> >
> > -* RTE_PKTMBUF_HEADROOM should be defined larger than sizeof(struct
> > virtio_net_hdr), which is 10 bytes.
> > +* RTE_PKTMBUF_HEADROOM should be defined
> > + larger than sizeof(struct virtio_net_hdr), which is 10 bytes,
> > +when
> > using non-mergeable.
> > + larger than sizeof(struct virtio_net_hdr_mrg_rxbuf), which is 12
> > bytes when mergeable or F_VERSION_1
> > + is set.
> >
>
> All constants and struct names should be ```` quoted. Like this:
>
>
> * ``RTE_PKTMBUF_HEADROOM`` should be defined larger than
> ``sizeof(struct virtio_net_hdr)``, which is 10 bytes, when using
> non-mergeable or larger than ``sizeof(struct virtio_net_hdr_mrg_rxbuf)``,
> which is 12 bytes when mergeable or ``F_VERSION_1`` is set.
>
Good suggestion.
Zhiyong Yang
> John
>
@@ -41,8 +41,8 @@ The DPDK extends kni to support vhost raw socket interface,
which enables vhost to directly read/ write packets from/to a physical port.
With this enhancement, virtio could achieve quite promising performance.
-In future release, we will also make enhancement to vhost backend,
-releasing peak performance of virtio PMD driver.
+In future release, we will also make enhancement to vhost backend.
+Peak performance of virtio PMD driver will be released with every release version since DPDK 17.02 release.
For basic qemu-KVM installation and other Intel EM poll mode driver in guest VM,
please refer to Chapter "Driver for VM Emulated Devices".
@@ -73,15 +73,20 @@ In this release, the virtio PMD driver provides the basic functionality of packe
* It supports multicast packets and promiscuous mode.
-* The descriptor number for the Rx/Tx queue is hard-coded to be 256 by qemu.
+* The descriptor number for the Rx/Tx queue is hard-coded to be 256 by qemu 2.7 and below.
If given a different descriptor number by the upper application,
the virtio PMD generates a warning and fall back to the hard-coded value.
+ Rx queue size can be configureable and up to 1024 since qemu 2.8 and above. Rx queue size is 256
+ by default. Tx queue size is still hard-coded to be 256.
* Features of mac/vlan filter are supported, negotiation with vhost/backend are needed to support them.
When backend can't support vlan filter, virtio app on guest should disable vlan filter to make sure
the virtio port is configured correctly. E.g. specify '--disable-hw-vlan' in testpmd command line.
-* RTE_PKTMBUF_HEADROOM should be defined larger than sizeof(struct virtio_net_hdr), which is 10 bytes.
+* RTE_PKTMBUF_HEADROOM should be defined
+ larger than sizeof(struct virtio_net_hdr), which is 10 bytes, when using non-mergeable.
+ larger than sizeof(struct virtio_net_hdr_mrg_rxbuf), which is 12 bytes when mergeable or F_VERSION_1
+ is set.
* Virtio does not support runtime configuration.